This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and

Transcription

1 This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier s archiving and manuscript policies are encouraged to visit:

2 Available online at Journal of Research in Personality 42 (2008) Pleasure and pressure based prosocial motivation: Divergent relations to subjective well-being q Jochen E. Gebauer a, *, Michael Riketta b, Philip Broemer c, Gregory R. Maio a a School of Psychology, Cardiff University, 70 Park Place, Cardiff, CF10 3AT, Wales, UK b Aston University, England, UK c University of Tübingen, Germany Available online 14 July 2007 Abstract We propose two fundamentally different motives for helping: gaining pleasure and fulfilling one s duty ( pressure ). Using the newly developed Pleasure and Pressure based Prosocial Motivation Scale, we demonstrated the distinctiveness of pleasure and pressure based prosocial motivation in three studies. Although the two motives exhibited different relations to a variety of personality characteristics, they were similarly related to trans-situational helping. Of particular interest, pleasure based prosocial motivation was positively related to self-actualization, self-esteem, life satisfaction, and positive affect and negatively related to negative affect. On the contrary, pressure based prosocial motivation was unrelated to self-actualization, self-esteem, life satisfaction, and positive affect but positively related to negative affect. These results qualify research showing that prosocial life goals generally increase subjective well-being. Ó 2007 Elsevier Inc. All rights reserved. Keywords: Prosocial motivation; Prosocial personality; Altruism; Pleasure; Pressure; Duty q We thank Katharina Noebels, Frank Nonnenmacher, and Johannes Mander for stimulating discussions on the motives underlying the prosocial personality. We also thank Laura King, Christina Howard, Kerry Rees, and two anonymous reviewers for their helpful comments on an earlier draft of this paper. Further, we are grateful to John Krantz and Ulf Reips for advertising our online-studies on their web portals. * Corresponding author. address: mail@jochengebauer.info (J.E. Gebauer) /$ - see front matter Ó 2007 Elsevier Inc. All rights reserved. doi: /j.jrp

3 400 J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) Introduction Helpful behavior depends on situational variables (e.g., Batson, 1991; Latané & Darley, 1970; Piliavin, Dovidio, Gaertner, & Clark, 1981) and on stable dispositions (e.g., Bierhoff, Klein, & Kramp, 1991; Eisenberg et al., 2002; Penner & Finkelstein, 1998; Staub, 1974). Research on both aspects has raised important questions about why people help, with a large amount of debate about whether helpfulness is always motivated by egoistic concerns (e.g., Cialdini, Brown, Lewis, Luce, & Neuberg, 1997) or whether helpfulness can also be an expression of pure selflessness (e.g., Batson, 1998). The present research takes a different perspective on this issue. Specifically, we postulate two broad motives underlying a prosocial orientation as a stable personality variable (i.e., the prosocial personality): Pleasure based prosocial motivation and pressure based prosocial motivation. In this paper, we introduce a measure of these motives and show that the distinction between the two motivations provides a more complete understanding of the prosocial personality Pleasure and pressure based prosocial motivation The most frequently mentioned motive in the literature on the motives that underlie the prosocial personality is the motive to gain pleasure from helping. The possibility that a prosocial orientation is motivated by the anticipation of positive affect (pleasure) is acknowledged by virtually all scholars in the field (e.g., Archer, Diaz-Loving, Gollwitzer, Davis, & Foushee, 1981; Batson, 1987; Baumann, Cialdini, & Kenrick, 1981; Cialdini et al., 1997; Harris, 1977; Schaller & Cialdini, 1988; Smith, Keating, & Stotland, 1989; Weiss, Buchanan, Alstatt, & Lombardo, 1971; Williamson & Clark, 1989). Following this research, we call the first motivation that underlies the prosocial personality pleasure based prosocial motivation. Importantly, the motivation to help others in order to gain pleasure can be contrasted with the motivation to fulfill a duty or conform to a social norm what we call pressure based prosocial motivation. For example, Campbell (1975) argues that helping behavior is motivated by pressure from cultural institutions, Rosenhan (1970) proposed a concept called normative altruism, and Bierhoff et al. (1991) found that people who help in the event of an accident possess a profound sense of duty. It is of interest that, there are several broader theories on motivation indicating that it may be crucial to distinguish between pleasure and pressure based prosocial motivation. First, the distinction between pleasure and pressure based prosocial motivation is consistent with the long-standing distinction between intrinsic and extrinsic motivation, respectively (e.g., Deci & Ryan, 1985). We conceptualize pleasure based prosocial motivation as more of an intrinsically driven motivation, whereas we conceptualize pressure based prosocial motivation as more extrinsically driven. In addition, Higgins (e.g., 1987, 1989) distinguishes between goals that function as ideals and goals that function as oughts. Ideals are the representation of attributes you would ideally like to possess, whereas oughts are the representation of attributes you feel you should or ought to possess. Further, Higgins (1997) proposed that ideals are associated with a promotion focus (a subjective framing of one s own behavior as approaching), whereas oughts are associated with a prevention focus (a subjective framing of one s own behavior as avoiding). Pleasure based prosocial motivation is conceptualized as involving ideal representations, whereas pressure based prosocial motivation is conceptualized as involving ought representations. Consequently,

4 pleasure and pressure based prosocial motivation are conceptualized to be associated with higher promotion and prevention foci, respectively. Finally, the even broader motivational distinction between seeking pleasure and avoiding pain (e.g., Freud, 1920) is another indicator that it may be crucial to distinguish between pleasure and pressure based prosocial motivation. That is, pleasure based prosocial motivation may be a specific instantiation of the search for pleasure, whereas pressure based prosocial motivation may be a specific instantiation of avoiding guilt due to a lack of fulfillment of one s perceived duties. Although these models of motivation support our claim that pleasure and pressure based prosocial motivation correspond to fundamental dimensions of motivation, our main reason for focusing on these two particular motives is that they may help to explain inconsistent findings in the literature concerning the relations between prosocial motives and subjective well-being. Prosocial goals and values have been assumed to facilitate the satisfaction of fundamental psychological needs, such as relatedness and growth (Deci & Ryan, 1985; Fromm, 1976; Maslow, 1954; Rogers, 1961). The satisfaction of fundamental psychological needs, in turn, should lead to higher subjective well-being (Deci, 1980; Maslow, 1954). Consequently, many researchers have concluded that a prosocial orientation should positively contribute to subjective well-being (see Kasser, 2000, for a review), and many tests have provided evidence that other-oriented life goals are indeed positively related to subjective well-being (Kasser & Ryan, 1993, 1996, 2001; Kasser, Ryan, Zax, & Sameroff, 1995; Ryan, Sheldon, Kasser, & Deci, 1996; Schmuck, 2001; Sheldon & Kasser, 1998). However, Sagiv and Schwartz (2000) failed to find this positive relationship in a large-scale study that tested the relation between subjective well-being and an interpersonal value orientation, which included helpfulness as one component (see Schwartz, 1992). A possible reason for these divergent findings may be that it is crucial to distinguish between pleasure and pressure based prosocial motivation. We showed above that a prosocial orientation is not necessarily intrinsically motivated. Ryan, Deci, and Grolnick (1995) provided evidence that only an intrinsic motivation to achieve one s goals, but not an extrinsic motivation to achieve one s goals, is positively associated with subjective well-being. Deci and Ryan (1995) found similar effects of intrinsic and extrinsic motivation to achieve one s goals on self-esteem. Similarly, if a prosocial orientation arises from a sense of duty or pressure, this orientation should not facilitate the satisfaction of innate psychological needs and thus should not cause higher subjective well-being. Indeed, Maslow (1954) assumed that commitment to altruism is a characteristic of self-actualizing people. As a consequence of this relation, we expected a positive relationship between pleasure based prosocial motivation and subjective well-being, which should be mediated by self-actualization. Thus, the second major aim of the current research was to test the hypothesis that pleasure based prosocial motivation, but not pressure based prosocial motivation, is positively related to subjective well-being. Support for this hypothesis would provide an explanation for the incoherent findings in the literature that investigates the relationship between prosocial motives and subjective well-being. 2. Study 1 J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) In Study 1, we tested the Pleasure and Pressure Based Prosocial Motivation Scale (3PMS), a new self-report measure that consists of two subscales measuring pleasure and pressure based prosocial motivation. The first aim of Study 1 was to show that the

5 402 J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) PMS possesses good psychometric properties. Specifically, we expected that the items of the 3PMS cluster into the two proposed dimensions and that the two subscales possess sufficient internal consistency. The second aim was to show that both subscales are positively related to trans-situational helping. To measure this trait, we used the most widely used measure to assess the prosocial personality, the Rushton Altruism Scale (RAS; Rushton, Chrisjohn, & Fekken, 1981). The RAS asks respondents to indicate the frequency with which they have performed each of 20 listed behaviors related to helpfulness. Rushton et al. (1981) validated the scale using a number of different criteria (e.g., peer reports) and reported internal consistencies ranging from a =.78 to a =.87. As a third aim, we tested the convergent and discriminant validity of the new scale using a broad set of measures that are conceptually related to pleasure and pressure based prosocial motivation. Our predictions were based on the assumption that pleasure based prosocial motivation is a more intrinsic motive, whereas pressure based prosocial motivation is more extrinsic. For example, we expected that pleasure based prosocial motivation, but not pressure based prosocial motivation, would be associated with higher levels of subjective-well being, because internal motivations are presumed to cause greater self-actualization, which is related to subjective well-being (e.g., Maslow, 1962, 1970). In addition, we expected that people who are high in pressure based prosocial motivation exhibit higher levels of personal need for structure and preference for consistency. These constructs reflect desires for internal cognitive consistency, which are maximized when people behave in accordance with internal values, beliefs, and attitudes. Duty or pressure in general can reflect such a commitment, in addition to reflecting external obligations (cf. Trafimow & Trafimow, 1999). Consistent with this view, Heine, Proulx, and Vohs (2006) recently argued that acting in line with moral norms is a way to satisfy one s need for consistency and structure. Thus, people with strong needs for consistency and structure should be highly motivated to fulfill moral duties, such as helpfulness, to satisfy their needs for consistency and structure (for a review see Cross, Gore, & Morris, 2003). Finally, we assessed the criterion validity of the Pleasure based Prosocial Motivation subscale by examining its ability to predict self-reported pleasure experienced directly after helpful behavior. Because the studies reported in this paper were conducted online and were not financially rewarded, participation in our studies was itself a helpful behavior. To make this aspect salient, we informed our participants in Study 1 that the research project was being conducted by an undergraduate student, who could not afford the money to pay participants, which was actually the case. We hypothesized that participants high in pleasure based prosocial motivation experience more enjoyment of the study, whereas pressure based prosocial motivation is unrelated to study enjoyment Method Participants Five hundred and ninety two (441 women, 138 men, and 13 unknown) participants completed this online-study ( The study was advertised on John Krantz s web portal for online-studies Psychological Research on the Net ( The mean age of the participants was years (SD = 9.41). Four hundred and fifty two participants were from North

6 J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) Table 1 Descriptive statistics and internal consistencies of Studies 1, 2, and 3 Measure M SD a Study 1 Prosocial personality Pleasure based prosocial motivation Pressure based prosocial motivation Self-actualization Self-esteem Positive affect Negative affect Preference for consistency Personal need for structure Study enjoyment Study 2 Prosocial personality Pleasure based prosocial motivation Pressure based prosocial motivation Life-satisfaction Interdependent self-construal Independent self-construal Study 3 Prosocial personality Pleasure based prosocial motivation Pressure based prosocial motivation Attachment anxiety Attachment avoidance America, 50 from Europe, 16 from Asia, 10 from Australia, 9 from Middle or South America, 3 from Africa, and 36 from elsewhere on the world (16 unknown). One hundred and fifty eight participants possessed a college or university degree, 421 did not possess such a degree, and 13 did not indicate whether or not they possessed a degree. Five hundred and sixteen participants indicated that English is their first language, 63 indicated that English is not their first language, and 13 did not respond to this item Procedure The language of the study was English. Participants completed questionnaires assessing self-esteem, preference for consistency, personal need for structure, subjective wellbeing, self-actualization, the prosocial personality, pleasure and pressure based prosocial motivation, and study enjoyment, in the listed order. Participants then read a feedback letter and were thanked for their participation. The descriptive statistics and internal consistencies of the measures are reported in Table 1. The internal consistencies were satisfactory Self-esteem. The Self-Liking/Self-Competence Scale Revised Version (Tafarodi & Swann, 2001) contains an 8-item Self-Liking subscale and an 8-item Self-Competence subscale. Example items for the Self-Liking subscale are I am very comfortable with myself and I feel great about who I am. Example items for the Self-Competence subscale are I am highly effective at the things I do and I am very talented. Participants

7 404 J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) responded to each item using a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). As in previous research (e.g., Bernard, Gebauer, & Maio, 2006; Verplanken, Friborg, Wang, Trafimow, & Woolf, 2007), the two subscales were averaged to create a global self-esteem score Preference for consistency. The short form of the Preference for Consistency Scale (Cialdini, Trost, & Newsom, 1995) contains 9 items. Example items are It is important to me that those who know me can predict what I will do and I make an effort to appear consistent to others. Participants responded to each item using a 5-point scale ranging from 1 (strongly disagree) to 5(strongly agree) Personal need for structure. The Personal Need for Structure Scale (Thompson, Naccarato & Parker, 1992, as reported in Neuberg & Newsom, 1993) contains 12 items. Example items are I enjoy having a clear and structured mode of life and I become uncomfortable when the rules in a situation are not clear. Participants responded to each item using a 5-point scale ranging from 1 (strongly disagree) to 5(strongly agree) Subjective well-being. The PANAS (Watson, Clark, & Tellegen, 1988) contains a 10-item Positive Affect subscale and a 10-item Negative Affect subscale. Example items for the Positive Affect subscale are enthusiastic and active. Example items for the Negative Affect subscale are upset and ashamed. Participants indicated whether or not I generally feel this way... using a 5-point scale ranging from 1 (very slightly or not at all) to 5 (extremely) Self-actualization. The Self-Actualization Index (Jones & Crandall, 1986) contains 15 items, such as It is better to be yourself than to be popular and I can express my feelings even when they may result in undesirable consequences. Participants responded to each item using a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree) Prosocial personality. The Rushton Altruism Scale (Rushton et al., 1981) contains 20 items, such as I have given money to a charity and I have helped carry a stranger s belongings (books, parcels, etc.). Participants responded to each item using a 5- point scale ranging from 1 (never) to 5(very often) Pleasure and pressure based prosocial motivation. The Pleasure and Pressure based Prosocial Motivation Scale (3PMS) contains a 4-item Pleasure based Prosocial Motivation subscale and a 4-item Pressure based Prosocial Motivation subscale with two reverse-scored items per subscale. The items were formulated for this study. Participants responded to each item using a 5-point scale ranging from 1 (strongly disagree) to 5(strongly agree). The items are listed in Table 2. The items were presented in the same order as shown in Table Study enjoyment. To assess the pleasure caused by participating in the current study, we asked participants, Did you enjoy this study? Participants responded to this item using a 5-point scale ranging from 1 (not at all) to 5(very much). Most participants reported moderate amounts of pleasure from participation (see Table 1).

8 J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) Table 2 Items of the Pleasure and Pressure based Prosocial Motivation Scale (3PMS) Subscale Item Pleasure Supporting other people makes me very happy. Pressure I do not feel obligated to perform selfless acts towards others. (R) Pleasure I do not have a great feeling of happiness when I have acted unselfishly. (R) Pressure I feel indebted to stand up for other people. Pleasure When I was able to help other people, I always felt good afterwards. Pressure I do not regard it as my duty to act selflessly. (R) Pleasure Helping people who are doing not well does not raise my own mood. (R) Pressure I feel a strong duty to help other people in every situation where it is possible for me. Note. Pleasure, pleasure based prosocial motivation; pressure, pressure based prosocial motivation; R, reverse scoring Results and discussion Zero-order correlations The zero-order correlations between all measures are reported in Table 3. The inter-correlations between the measures were small to moderate. Table 3 Inter-correlations between all measures used in Studies 1, 2, and 3 (1) (2) (3) (4) (5) (6) (7) (8) (9) Study 1 (1) Prosocial personality (2) Pleasure motivation.45 ** (3) Pressure motivation.39 **.58 ** (4) Self-actualization.48 **.60 **.40 ** (5) Self-esteem.25 **.24 **.11 **.50 ** (6) Positive affect.39 **.37 **.29 **.46 **.42 ** (7) Negative affect.09*.08 ns.22 **.06 ns.35 **.32 ** (8) Pref. for consistency.15 **.15 **.24 **.02 ns.14 **.20 **.12 ** (9) Need for structure.08 ns.18 **.24 **.08*.05 ns.14 **.27 **.58 ** (10) Study enjoyment.26 **.30 **.23 **.31 **.13 **.26 **.07 ns.16 **.13 ** Study 2 (1) Prosocial personality (2) Pleasure motivation.37 ** (3) Pressure motivation.34 **.31 ** (4) Life satisfaction.06 ns.23 **.04 ns (5) Interdependence.19*.25*.30 **.07 ns (6) Independence.06 ns.11 ns.15 ns.35 **.09 ns Study 3 (1) Prosocial personality (2) Pleasure motivation.58 ** (3) Pressure motivation.44 **.43 ** (4) Attachment anxiety.11 ns.29*.16 ns (5) Attachment avoidance.19 ns.27*.02 ns.12 ns Note. ** p <.01, * p <.05, ns p >.05.

9 406 J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) Psychometric properties Our first aim was to show that a two-dimensional model of the motivation underlying a prosocial orientation (pleasure and pressure) fits the data. As expected, a confirmatory factor analysis revealed that our two factor model fits the data well, CFI =.95; RMR =.08; CMIN = 83.07, and revealed a significantly better fit than a one factor solution (chi 2 = , p <.0001). 1 The two factors correlated at r =.52, p < Furthermore, the internal consistencies of the two subscales (a [Altruism Pleasure] =.80, a [Altruism Pressure] =.70) were good Relation to the prosocial personality We conducted a multiple regression analysis to examine the amount of variance of the prosocial personality that was uniquely predicted by pleasure and by pressure based prosocial motivation. The upper part of Table 2 shows the results of this analysis. As expected, pleasure and pressure based prosocial motivation were significant independent predictors of increased trans-situational helping (as indexed by the RAS) Convergent and discriminant validity We conducted a series of multiple regression analyses to examine the extent to which pleasure and pressure based prosocial motivation uniquely explained variance in each of the following criterion variables: (a) self-actualization, (b) self-esteem, (c) positive affect, (d) negative affect, (e) preference for consistency, (f) personal need for structure, and (g) study enjoyment. As can be seen in Table 4, pleasure based, but not pressure based, prosocial motivation was positively related to self-actualization, self-esteem, and positive affect and it was negatively related to negative affect. In contrast, pressure based, but not pleasure based, prosocial motivation was positively related to negative affect and needs for consistency and structure. Finally, pleasure based, but not pressure based, prosocial motivation was positively related to enjoying participation in the study. Our prediction of the positive relationship between pleasure based prosocial motivation and higher subjective well-being as well as higher self-esteem was based on the hypothesis that pleasure based prosocial motivation affects self-actualization, which in turn should increase subjective well-being and self-esteem. We tested this assumption using the mediation analysis recommended by Baron and Kenny (1986). Necessary conditions for mediation are that the independent variable (pleasure based prosocial motivation) correlates with the mediator (self-actualization) and that the mediator correlates with the outcome 1 In this model, we expected the error variances of the two reverse-scored as well as the two non-reverse scored items of each factor to be related because of common method variance (Marsh, 1996). 2 The correlation between the two subscales was not theoretically expected but was expected to occur based on methodological issues. First, given that the items of the two subscales were intermixed, it is likely that carryover effects occurred. Second, participants may have tried to be as consistent in their answers as possible because they may have incorrectly perceived the scale as a one-dimensional scale. Third, participants responses may have been partially based on self-perception (Bem, 1967; Bem, 1972). That is, participants may have, to a certain extend, reasoned that they must be motivated to help others by seeking pleasure and by feeling pressure because they show helpful behaviour. The positive relation between the subscales is however not problematic at all. First, the relation is only moderate, and second, when testing the relation between one motivation and its correlate we control for the other motivation and thus partial out the method based overlap between the two motivations (for identical issues and identical solution to these issues, see Brennan, Clark, & Shaver, 1998; Tafarodi & Swann, 2001).

10 J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) Table 4 Standardized regression coefficients (bs) of pleasure and pressure based prosocial motivation on the criterion variables of Studies 1, 2, and 3 Criterion Predictors Pleasure Pressure Study 1 Prosocial personality.34 ***.19 *** Self-actualization.56 ***.07 ns Self-esteem.27 ***.05 ns Positive affect (controlling for Negative affect).32 ***.04 ns Negative affect (controlling for Positive affect).18 ***.23 *** Preference for consistency.01 ns.23 *** Personal need for structure.06 ns.20 *** Study enjoyment.26 ***.07 ns Study 2 Prosocial personality.29 ***.25 ** Life satisfaction.27 **.13 ns Interdependent self-construal (controlling for independent self-construal).18 *.24 ** Independent self-construal (controlling for interdependent self-construal).06 ns.13 ns Study 3 Prosocial personality.48 ***.24 * Attachment anxiety (controlling for attachment avoidance).23 *.11 ns Attachment avoidance (controlling for attachment anxiety).22 *.04 ns Note. The predictors were entered simultaneously in the regression. Pleasure, pleasure based prosocial motivation; pressure, pressure based prosocial motivation. * p <.05. ** p <.01. *** p <.001. ns p >.05. variable (subjective well-being or self-esteem). As shown in Table 1, these conditions were met. An additional condition is that the relationship between the independent variable and the outcome variable is significantly reduced when the mediator is controlled. Satisfying this condition, when pleasure based prosocial motivation and self-actualization were simultaneously regressed on subjective well-being or self-esteem, the effects of pleasure based prosocial motivation on positive affect, negative affect, and self-esteem were nonsignificant, b(591) =.09, ns, b(591) =.01, ns,andb(591) =.04, ns, respectively. The Sobel test (Sobel, 1982) revealed that these relationships were significantly weaker than the zeroorder effects for positive affect, z = 6.86, p <.001, for negative affect, z = 3.66, p <.001, and for self-esteem, z = 9.31, p <.001. In contrast, the effects of pleasure based prosocial motivation on self-actualization remained significant after controlling for positive affect, b(591) =.48, p <.001, for negative affect, b(591) =.55, p <.001, and for self-esteem, b(591) =.36, p <.001. Thus, these analyses supported the assumption that pleasure based prosocial motivation affects self-esteem and subjective well-being through its effect on selfactualization. We wanted to conduct one final set of analyses to test an alternative explanation for the relation between pleasure based prosocial motivation and (a) self-actualization, (b) self-esteem, and (c) study enjoyment. Specifically, it is possible that the relations between pleasure based prosocial motivation and these variables are not due to a positive effect of

11 408 J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) pleasure based prosocial motivation on these variables, but instead that the relations are actually spurious ones, which are caused by positive affect. In fact, the mood-as-information hypothesis (Schwarz & Clore, 1983) predicts that one source of information on which people base their self-evaluations is their current mood. That is, people in a positive mood should judge themselves as more self-actualized, as higher in self-esteem, as enjoying the study more, and also as enjoying helping in general more. In line with the mood-as-information hypothesis, we found that positive affect related positively to self-actualization, self-esteem, and study enjoyment (see Table 3). However, does positive affect completely account for the relationship between pleasure based prosocial motivation and its criterion variables? To test this alternative hypothesis, we conducted three multiple regression analyses with pleasure based prosocial motivation as the predictor and (a) self-actualization, (b) self-esteem, and (c) study enjoyment as the criterion variables, respectively. Crucially, in all three analyses we controlled for positive affect and pressure based prosocial motivation. The results of the multiple regression analyses do not support the alternative hypothesis. In fact, the effects of pleasure based prosocial motivation on self-actualization, b(591) =.48, p <.001, self-esteem, b(591) =.15, p <.001, and study enjoyment, b(591) =.21, p <.001, all remained highly significant Summary People high in trans-situational helping (as indexed by the RAS) exhibit higher levels of both pleasure and pressure based prosocial orientation. This result supports our hypothesis that both motives underlie the prosocial personality. The data also support our hypothesis that prosocial goals and values are not necessarily associated with high subjective well-being and self-esteem. As expected, only pleasure but not pressure based prosocial motivation, was positively associated with subjective well-being and selfesteem. In addition, participants who were high in pressure based prosocial motivation exhibited more negative affect and higher needs for consistency and structure. Furthermore, as predicted, only pleasure but not pressure based prosocial motivation was positively associated with study enjoyment. Moreover, we found that self-actualization completely mediated the relationship between pleasure based prosocial motivation and subjective well-being as well as self-esteem. These results are consistent with Maslow s (1962, 1970) assumption that intrinsic motivation leads to higher levels of self-actualization, which leads to higher subjective well-being. Finally, we provided evidence against the alternative explanation that the relations between pleasure based prosocial motivation and (a) self-actualization, (b) self-esteem, and (c) study enjoyment are spurious ones, which are due to an effect of positive affect on all of these constructs (cf. Schwarz & Clore, 1983). 3. Study 2 Study 2 addressed three goals. The first goal was to replicate the finding of Study 1 that both prosocial motivations predict the prosocial personality. The second goal was to further investigate the divergent relationship between the two prosocial motivations and subjective well-being. Diener (1984) distinguished between an affective and a cognitive component of subjective well-being. In Study 1, we focused solely on the affective component by assessing subjective well-being with the PANAS. In Study 2, we therefore decided

12 to focus on the cognitive component of subjective well-being, namely life satisfaction (e.g., Diener & Diener, 1995). Study 2 also examined a potential cause of pleasure and pressure based prosocial motivation. People with a strong interdependent self-construal perceive themselves primarily in relation to other people (e.g., Markus & Kitayama, 1991). Helping other people should be perceived as both a pleasure and a duty for people with an interdependent self-construal. On the one hand, they should regard helping as a pleasure because it fosters feelings of closeness to others. On the other hand, they should regard helping as a duty because people with an interdependent self-construal in general emphasize norms and duties as guiding principles for their behavior (e.g., Lee, Aaker, & Gardner, 2000; Triandis, 1995). In contrast, it is more difficult to predict a clear pattern of relations between pleasure and pressure based prosocial motivation and an independent self-construal, which is the tendency to perceive oneself primarily as an individual that is independent from or different to other people (e.g., Markus & Kitayama, 1991). The reason for this difficulty is that there are theoretical reasons to expect a negative as well as a positive relationship between pleasure and pressure based prosocial motivation and an independent self-construal. On the one hand, an independent self-construal is associated with a stronger concern for the self (i.e., self-promotion rather than other-promotion; e.g., Markus & Kitayama, 1991). Thus, people with a strong independent self-construal may neither gain pleasure from helping others, nor may they feel any duty to help others. On the other hand, given that helpfulness is at least in many cases associated with egoistic concerns (e.g., Cialdini et al., 1997), people with a strong independent self-construal may be motivated to help in situations where helping promotes the self. Following this rationale, a positive relationship between an independent self-construal and our prosocial motivations would be expected. Taken together, we did not formulate a specific hypothesis concerning the relationship between our prosocial motivations and an independent self-construal. However, our theoretical analysis makes it appear likely that the expected opposing effects result in null relations between the two prosocial motivations and an independent self-construal Method J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) Participants One hundred and fifty-five (109 women and 46 men) participants completed this onlinestudy ( The study was advertised on Ulf Reips s web portal for online-studies Web Experimental Psychology Lab ( The mean age of the participants was years (SD = 8.82). No data about nation of origin, education, and language were collected for this study Procedure The language of the study was German. Participants completed questionnaires assessing life satisfaction, the prosocial personality, pleasure and pressure based prosocial motivation, and self-construal, in the listed order. Participants then read a feedback letter and were thanked for their participation. The measures of pleasure and pressure based prosocial motivation were the same as in Study 1. The remaining measures are described below. The descriptive statistics and internal consistencies of the measures are reported in Table 1.

13 410 J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) Inspection of the table reveals that most of the internal consistencies were satisfactory or good Life satisfaction. The Satisfaction with Life Scale (Diener, Emmons, Larsen, & Griffin, 1985) contains 5 items. Example items are In most ways my life is close to my ideal and If I could live my life over, I would change almost nothing. Participants responded to each item using a 7-point scale ranging from 1 (applies not at all) to 7(applies completely) Prosocial personality. The Prosocial Motivation Questionnaire (Claar, Boehnke, & Silbereisen, 1984) was designed and validated for assessing a prosocial orientation in German and Polish high school students. 4 We used a short form of the scale by deleting those items that can only be applied to high school students. Example items from the remaining 9-item short form are I never give money to charity collections on the street (reverse-scored) and I refuse to help people I don t know (reverse-scored). Participants responded to each item using a 7-point scale ranging from 1 (applies not at all)to7(applies completely) Self-construal. The Independent-Interdependent Self-Construal Scale (Singelis, 1994) contains a 12-item Independent Self-Construal subscale and a 12-item Interdependent Self-Construal subscale. Example items for the Independent Self-Construal subscale are I am comfortable with being singled out for praise or rewards and I act the same way no matter who I am with. Example items for the Interdependent Self-Construal subscale are I have respect for the authority figures with whom I interact and My happiness depends on the happiness of those around me. Participants responded to each item using a 7-point scale ranging from 1 (applies not at all) to 7(applies completely) Results and discussion Zero-order correlations The zero-order correlations between all measures are reported in Table 3. The inter-correlations between the measures were small to moderate Independent effects of pleasure and pressure based prosocial motivation We conducted a series of multiple regression analyses to examine the extent to which pleasure and pressure based prosocial motivation uniquely predict the following criterion variables: (a) the prosocial personality, (b) subjective well-being, and (c) self-construal. The results are shown in Table 4 and described below. 3 The internal consistency of the Pressure based Prosocial Motivation subscale of the 3PMS is rather low in this sample. However, it is adequate in Study 3, which uses the same German version of the 3PMS. Taken Studies 2 and 3 together, the internal consistency of the 3PMS is a =.65, which is comparable to the internal reliability of the English version of the Pressure based Prosocial Motivation subscale of the 3PMS (a =.70). Comparability of the internal consistencies is an important indicator of the similarity of the two translations of the 3PMS. Further, it speaks for the suitability of the 3PMS in both languages that the internal consistencies of the Pleasure based Prosocial Motivation subscales are similar (both a =.80). Finally, the fact that we found similar relations between the 3PMS and the prosocial personality and subjective well-being across both languages gives further credence to the suitability of both translations of the 3PMS. 4 Despite its title, none of the items actually pertain to motivation per se.

14 Replication of relation with the prosocial personality. A multiple regression analysis examined the unique effects of pleasure and pressure based prosocial motivation on the prosocial personality. Consistent with Study 1, both motivations were significant independent predictors of the prosocial personality, despite the use of a different measure of the prosocial personality and the different cultural context (see Table 4) Relation to subjective well-being. We used a similar regression analysis to examine the ability of pleasure and pressure based prosocial motivation to independently predict life satisfaction. Only pleasure but not pressure based prosocial motivation was positively related to life satisfaction (see Table 4). This result replicates Study 1 for a cognitive, rather than an affective, component of subjective well-being Relation to self-construal. As shown in Table 4, we found that both pleasure and pressure based prosocial motivation were positively related to an interdependent self-construal and unrelated to an independent self-construal. 4. Study 3 Study 3 further examines the role of significant others in the motivations underlying the prosocial personality. Recently, Mikulincer, Shaver, and colleagues (Gillath, Shaver, & Mikulincer, 2005; Gillath et al., 2005; Mikulincer, Shaver, Gillath, & Nitzberg, 2005; for a review see Mikulincer & Shaver, 2005) have repeatedly demonstrated that secure attachment increases helpfulness. Attachment theory (Bowlby, 1982/1969) assumes that early childhood interactions result in two types of mental representations, called working model of the self (or attachment anxiety; see Brennan et al., 1998) and working model of others (or attachment avoidance; see Brennan et al., 1998). Secure attachment is characterized by the simultaneous presence of low attachment anxiety and low attachment avoidance (i.e., positive working models of the self and others). As such, Mikulincer and Shaver (2005) have argued that securely attached people should have more resources available to help others than people who are insecurely attached. This should simply be the case because securely attached people are less anxious and avoidant and thus should be less concerned about their own security. In this vein, it is interesting to test whether secure attachment is associated with pleasure and pressure based prosocial motivation. Because securely attached people have positive emotions toward other people, securely attached people should gain pleasure from helping others. Therefore, we hypothesize a positive relationship between secure attachment and pleasure based prosocial motivation. However, there is no reason to assume that securely attached people should feel more or less duty to help others. Thus, we hypothesize no relationship between secure attachment and pressure based prosocial motivation Method J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) Participants Eighty (57 women and 23 men) participants completed this online-study ( The study was also advertised on Ulf Reips s web portal for online-studies Web Experimental Psychology Lab. The mean age of the participants was years (SD = 7.83).

15 412 J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) Procedure The language of the study was German. Participants completed questionnaires assessing attachment, the prosocial personality, and pleasure and pressure based prosocial motivation, in the listed order. Finally, participants read a feedback letter and were thanked for their participation. The measures of the prosocial personality, and pleasure and pressure based prosocial motivation were the same as in Study 2. The measure of attachment is described below. The descriptive statistics and internal consistencies of the measures are reported in Table 1. Most of the internal consistencies were satisfactory Attachment. The Experiences in Close Relationships Questionnaire (Brennan et al., 1998) contains an 18-item subscale assessing attachment anxiety and an 18-item subscale assessing attachment avoidance. Example items for the Attachment Anxiety subscale are I worry about being abandoned and I worry a lot about my relationships. Example items for the Attachment Avoidance subscale are I prefer not to be too close to romantic partners and I prefer not to show a partner how I feel deep down. Participants responded to each item using a 7-point scale ranging from 1 (applies not at all) to 7(applies completely). Secure attachment is characterized by the simultaneous presence of low scores of attachment anxiety and attachment avoidance (Brennan et al., 1998) Results and discussion Zero-order correlations The zero-order correlations between all measures are reported in Table 3. The inter-correlations between the measures were small to moderate Independent effects of pleasure and pressure based prosocial motivation We conducted two multiple regression analyses to examine the extent to which pleasure and pressure based prosocial motivation uniquely predict the following criterion variables: (a) the prosocial personality, (b) secure attachment. The results are shown in Table 4 and described below Replication of relation with the prosocial personality. A multiple regression analysis examined the unique effects of pleasure and pressure based prosocial motivation on the prosocial personality. Consistent with Studies 1 and 2, both motives were significant independent predictors of the prosocial personality (see Table 3) Relation to secure attachment. Using the same regression procedure as above, we found that pleasure but not pressure based prosocial motivation was negatively related to attachment anxiety and avoidance. These results are consistent with the claim that pleasure but not pressure based prosocial motivation is influenced by secure attachment. 5. Gender and age differences in pleasure and pressure based prosocial motivation across studies The final goal of the present research was to explore gender and age differences in the two motivations to help. We decided to test gender and age differences in pleasure and pressure based prosocial motivation across all three studies because of the relatively small

16 number of male participants in each study. Analyzing the data from our three studies together resulted in a sample size of 812 participants (596 women, 203 men, and 13 unknown). The mean age of the participants was M = (SD = 9.27). We had no specific hypotheses concerning gender and age differences in the two motivations to help. A MANOVA with gender as the sole factor and pleasure and pressure based prosocial motivation as the two dependent variables revealed that women scored significantly higher than men on pleasure, F(2, 797) = 18.39, p <.001, and pressure, F(2, 797) = 6,43, p <.05, based prosocial motivation. Age was unrelated to pleasure, r(797) =.03, ns, and pressure, r(797) =.00, ns, based prosocial motivation. The finding that women scored higher on pleasure as well as on pressure based prosocial motivation is consistent with the prior literature on gender differences in helping behavior. Eagly (1987) has shown that common gender stereotypes portray women as more altruistic than men. Indeed, an empirical review of the relation between gender and empathy found that women are more empathic then men (Eisenberg & Lennon, 1983; Lennon & Eisenberg, 1987). Empathy, in turn, has been shown to be a strong predictor of prosocial behavior (e.g., Batson, 1991; Cialdini et al., 1997). Moreover, Bybee (1998) reported evidence that women score higher on guilt feelings than men. Given that a feeling of duty might arise from a generalized feeling of guilt, Bybee s findings suggest that women may also score higher on pressure based prosocial motivation than men. 6. General discussion J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) Although there is considerable evidence for the prosocial personality (e.g., Bierhoff et al., 1991; Eisenberg et al., 2002; Penner & Finkelstein, 1998; Staub, 1974), there have been relatively few analyses of the motivational underpinnings of this trait. We have proposed that different motivations underlie the prosocial personality and that these different motivations, in turn, are associated with different personality variables. In this context, we derived two general motivations underlying the prosocial personality by reviewing the literature on the more specific motivations to help. As explained in the Introduction, pleasure and pressure based prosocial motivation fit with more general distinctions between two types of self-guides (Higgins, 1987, 1989), self-regulatory orientations (promotion vs. prevention, Higgins, 1997), the distinction between intrinsic and extrinsic motivation (Deci & Ryan, 1985), and the distinction between seeking pleasure and avoiding pain (Freud, 1920). In Study 1, we confirmed the postulated dimensionality of our new measure of these motives and provided evidence for internal consistency and convergent and discriminant validity of each subscale. Specifically, we found that people who reported a high pleasure based prosocial motivation exhibited higher levels of self-actualization, selfesteem, positive affect, and lower levels of negative affect, whereas pressure based prosocial motivation was unrelated to self-actualization, self-esteem, and positive affect and it was positively related to negative affect. This pattern is consistent with the assumption that pleasure based prosocial motivation is more of an intrinsic (vs. extrinsic) motivation than pressure based prosocial motivation. Study 1 also showed that the relations between pleasure based prosocial motivation and subjective well-being and self-esteem were completely mediated by self-actualization, a finding that supports our hypothesis that pleasure based prosocial motivation is related to intrinsic motivation. These findings qualify past evidence that a prosocial orientation is necessarily positively associated with higher subjective wellbeing (Kasser & Ryan, 1993, 1996, 2001; Kasser et al., 1995; Ryan et al., 1996; Schmuck,

17 414 J.E. Gebauer et al. / Journal of Research in Personality 42 (2008) ; Sheldon & Kasser, 1998; see Kasser, 2000, for a review). Our findings indicate that this relation is caused by those individuals who help for reasons of pleasure, rather than pressure. Crucially, our finding that only pleasure but not pressure based prosocial motivation was positively related to subjective well-being may explain why some researchers failed to find a relation between a prosocial orientation and high subjective well-being. For example, Sagiv and Schwartz (2000) did not find a positive relationship between an interpersonal social value orientation (e.g., helpfulness) and subjective well-being. One reason for the inconsistencies between the findings on the relations between prosocial goals (Kasser & Ryan, 1993, 1996, 2001; Kasser et al., 1995; Ryan et al., 1996; Schmuck, 2001; Sheldon & Kasser, 1998; see Kasser, 2000, for a review) and prosocial values (Sagiv & Schwartz, 2000) may be that prosocial goals are more intrinsically-driven ideals, whereas prosocial values are more extrinsically-driven oughts. Indeed, Rees and Maio (2006) asked participants to rate the extent to which they hold diverse social values as oughts and ideals. Results indicated that values emphasizing openness to novelty (e.g., freedom) were more strongly held as ideals than as oughts, whereas this was not true for values emphasizing benevolence (e.g., helpfulness), despite being rated as similar in importance. The studies by Rees and Maio suggest that benevolence values functioned more as ideals only when they were held as the most important values, which was true only for the minority of participants. Studies 2 and 3 shed some light on possible determinants of pleasure and pressure based prosocial motivation. As expected, we found that people high in pleasure and pressure based prosocial motivation were more likely to possess an interdependent self-construal. Further, we found that more securely attached participants scored higher on pleasure but not on pressure based prosocial motivation. We expected this pattern of findings because of the more general principle that people tend to derive more satisfaction from being with others when they feel connected to them and like them. This principle is evident in abundant research on effects of commitment and satisfaction in close relationships (for a review see Fletcher, 2002) and effects of identification and cohesion in groups (e.g., Shah, Kruglanski, & Thompson, 1998). We understand the current studies as a mere start in exploring pleasure and pressure based prosocial motivation. Thus, they leave a number of questions unanswered, which suggest themselves as topics for future research. For example, we proposed above that empathy is a precondition for pleasure based prosocial motivation, but that empathy is unrelated to pressure based prosocial motivation. This hypothesis has not been tested yet, although the gender difference we obtained is consistent with this hypothesis. Further, more research is needed to better connect pleasure and pressure based prosocial motivation to existing constructs in the more specific helpfulness literature and to the well-established research in the broader motivation literature. That is, on the one hand, in the Introduction we have proposed that our measure does not only assess the motivation underlying very specific types of helpfulness behavior (e.g., volunteerism) but that it is broader in a sense and thus assesses prosocial motivation more generally. On the other hand, we have proposed above that our measure is conceptually related to broader motivational distinctions such as intrinsic vs. extrinsic motivation (Deci & Ryan, 1985), ideal vs. ought self-guides (Higgins, 1987, 1989), promotion vs. prevention foci (Higgins, 1997), and pleasure seeking vs. pain avoidance (Freud, 1920). Thus, research is needed that tests the relations between the 3PMS and more specific helpfulness measures like the Volunteer